The understanding and treatment of pathological anxiety have long been a prime concern in regard to mental health. Alterations in GABAA function from controls are known to occur in many anxiety disorders including panic disorder, epilepsy, hypersensitive behavior, phobias, schizophrenia, alcoholism, Anglemans syndrome and Rhetts syndrome, as well as effects which lead to/or complicate drug abuse. The 1,4-benzodiazepines, employed to treat anxiety disorders as well as sleep disorders exhibit anxiolytic, anticonvulsant, muscle relaxant/ataxic, sedative- hypnotic and amnestic effects. Despite the clinical effectiveness of these drugs, there is a need for selective anxiolytics and anticonvulsants which are devoid of the sedative-hypnotic, muscle-relaxant, ataxic and amnestic effects. Recently, with the aid of computer modeling and chemical synthesis, we have developed an orally active anxioselective anxiolytic (1). This ligand exhibits very poor affinity at the a1 receptor subtype of the BzR/GABAergic system and has a near perfect efficacy profile in oocytes [alpha1beta3gamma2 (no efficacy), alpha2beta3gamma2 (full agonist), alpha3beta3gamma2 (50% agonist), alpha5beta3gamma2 (10-15% agonist);no affinity at alpha4/alpha6 BzR/GABAA receptors]. This ligand is orally active as an anxiolytic at 1mg/kg but has no sedative-hypnotic, muscle relaxant, or ataxic effects in rodents even up to 300 mg/kg. It does not generalize to chlordiazepoxide in the discriminative stimulus paradigm, nor has it shown any amnestic effects. It is also active in primates in the conflict paradigm with no signs of muscle relaxation, or ataxia or sedation. The prodrug (2) of this anxiolytic agent exhibits a similar profile in rodents. These two anxiolytics serve as the lead compounds in this study directed toward the synthesis of alpha2 and alpha3 subtype selective ligands. Moreover, these agents should be long-lived, metabolically stable, water soluble, orally active anxiolytic agents for the clinic. In addition, a new series of S-enantiomers of these agents will be prepared. These fit the computer model perfectly, while the R-isomers should be inactive. It is hoped these R-enantiomers may be antagonists of the S-isomers. It is felt 1 and 2 will serve as the lead compounds for the construction of much better, fast acting, orally active anxiolytic agents devoid of the sedative, ataxic, muscle relaxant and amnestic side effects of classical benzodiazepines, as well as exhibit reduced (or no) abuse potential. The lead compounds are illustrated in Figure I, while the target compounds in this phase of the research are depicted in Figure 2 and Schemes 1 -5. In the second part of this research (Schemes 6 and 7) emphasis is on the development of agonists, antagonists and inverse agonists that bind to alpha5beta3gamma2 subtypes with >400 fold subtype selectivity. Synthesis and pharmacological evaluation of these subtype selective agents will permit the assignment of the correct physiological functions to alpha5 subtypes. This is of special importance here in regard to cognition/amnesia and other processes mediated by the hippocampus. A network of 12 collaborators has been established to work with us on receptor binding, efficacy and pharmacology to understand the fundamental basis of/and treat anxiety disorders, as well as memory-impairment including age associated memory deficits. This latter research may also have implications in Alzheimer's disease and schizophrenia.